i-robot
|
e14942017d
|
!31673 support tranpose with neg perm
Merge pull request !31673 from 范吉斌/transpose_neg
|
4 years ago |
i-robot
|
c1a9e313b0
|
!31281 gpu dynamic shape
Merge pull request !31281 from changzherui/gpu_dynamic_shape
|
4 years ago |
i-robot
|
ecde0749bb
|
!31576 dynamic shape CPU & GPU
Merge pull request !31576 from TuDouNi/dynamic_shape_gpu_cpu_2
|
4 years ago |
fan-jibin
|
37a658e4c2
|
support tranpose with neg perm
|
4 years ago |
changzherui
|
df3976d5a0
|
gpu dynamic shape
|
4 years ago |
ttudu
|
adb18b82ca
|
dynamic shape CPU & GPU
|
4 years ago |
i-robot
|
9c6fb2a67e
|
!31647 clean code
Merge pull request !31647 from baihuawei/cleancode1.7
|
4 years ago |
i-robot
|
27e417de6a
|
!31271 Reconstruct the GPU invoking mode
Merge pull request !31271 from zong_shuai/reconstruct_gpu_kernel
|
4 years ago |
baihuawei
|
d0f2974316
|
clean code
|
4 years ago |
i-robot
|
d3d17acdbc
|
!31601 Clean code
Merge pull request !31601 from VectorSL/fix-codex
|
4 years ago |
i-robot
|
faeeca1d3d
|
!31525 solve the bug of layer norm epsilon val is const
Merge pull request !31525 from zong_shuai/layer_norm_debug
|
4 years ago |
VectorSL
|
c1220ea054
|
fix codex
|
4 years ago |
zong-shuai
|
d42daf0dcb
|
reconstruct_gpu_kernel
|
4 years ago |
i-robot
|
dc5f77f585
|
!31298 build libmindspore_backend.so
Merge pull request !31298 from zhoufeng/backend-frontend-decoupling
|
4 years ago |
zong-shuai
|
8b89d5b739
|
debug
|
4 years ago |
i-robot
|
b7d45d7dd8
|
!31372 修复重复模块警告
Merge pull request !31372 from 王程浩/master
|
4 years ago |
zhoufeng
|
b7bb53ff73
|
build libmindspore_backend.so
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
|
4 years ago |
i-robot
|
3dd2aa584e
|
!31345 Clean code
Merge pull request !31345 from VectorSL/code_clean
|
4 years ago |
王程浩
|
d75525252e
|
fix duplication_code
|
4 years ago |
i-robot
|
f52272ae34
|
!31369 Fix code check 2.0 issue
Merge pull request !31369 from ZPaC/static
|
4 years ago |
i-robot
|
34f321d946
|
!31300 [DynamicShape][Gpu]support dynamic shape for the DCN network
Merge pull request !31300 from hanhuifeng/dyn_gpu_bugs
|
4 years ago |
ZPaC
|
42bdf2bc5b
|
Fix code check 2.0 issue
|
4 years ago |
i-robot
|
c9cb729cd3
|
!31130 fix bug of fake quant perchannel
Merge pull request !31130 from cjh9368/clean_static_error_master
|
4 years ago |
VectorSL
|
83882d4712
|
clean code
|
4 years ago |
hanhuifeng2020
|
089ce0386a
|
[DynamicShape][Gpu]support dynamic shape for the DCN network
|
4 years ago |
buxue
|
b7f90e2b16
|
implement Conv2dBackward and log of pool
|
4 years ago |
i-robot
|
4b15a46a49
|
!30858 Refactor cpu kernel factory.
Merge pull request !30858 from TronZhang/refactor_cpu_factory
|
4 years ago |
i-robot
|
f604a4e1a8
|
!31023 solve the bug of return err when input_shape less than 2 or more than 4 in mirror pad
Merge pull request !31023 from zong_shuai/mirror_pad
|
4 years ago |
TronZhang
|
db2931516e
|
refactor cpu factory
|
4 years ago |
cjh9368
|
1e4cde75f6
|
[MS][LITE] fix bug of fakequant perchannel
|
4 years ago |
i-robot
|
c04faad4ad
|
!30559 modify gpu opt adapter dynamic shape
Merge pull request !30559 from changzherui/mod_gpu_opt
|
4 years ago |
changzherui
|
47ef477fef
|
modify gpu opt adapter dynamic shape
|
4 years ago |
i-robot
|
df55f87899
|
!31029 fix attr length bug of Conv2DBackpropFilter
Merge pull request !31029 from zhangbuxue/fix_bug_of_conv2dbackpropfliter
|
4 years ago |
buxue
|
c24e9c3e93
|
fix bug of Conv2DBackpropFilter
|
4 years ago |
zhujingxuan
|
45e7c257af
|
remove duplicate op impl
matrix_triangular_solve_gpu_kernel and trsm_solve_gpu_kernel are actually the same.
|
4 years ago |
buxue
|
50f3ad00ab
|
develop maxpool3d, avgpool3d and conv3d on CPU(including forward and backward)
|
4 years ago |
zong-shuai
|
cdcbd9aeb0
|
debug
|
4 years ago |
ZPaC
|
04ee8f4dd9
|
Support reference for gpu and cpu for parameter update
|
4 years ago |
hanhuifeng2020
|
b571e67795
|
[DynamicShape][GPU]add dynamic shape support some ops for new network
|
4 years ago |
i-robot
|
ff49911889
|
!30572 delete CpuDynamicshape and GpuDynamicshape
Merge pull request !30572 from TuDouNi/gpu_cpu_delete_dynamic_kernel
|
4 years ago |
ttudu
|
dac6f5425d
|
delete GPU/CPUDynamicKernel
|
4 years ago |
Gaoxiong
|
8745e4d3f1
|
GraphKernel performance optimize: cache kernel addr in KernelMod
|
4 years ago |
ckey_Dou
|
e9679ca0bd
|
return default shape when max_shape is empty
|
4 years ago |
zhoufeng
|
f49b195c39
|
extract common as an independent shared library
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
|
4 years ago |
jinjiali
|
2c47b286dd
|
correct cublas path in cuda_ops cmakelists
|
4 years ago |
i-robot
|
b3c5943bf8
|
!30373 [DynamicShape][GPU]add dynamic shape support of Concat op and its backward for new Network
Merge pull request !30373 from hanhuifeng/dcn_dyn_ops_1
|
4 years ago |
i-robot
|
ac1463a192
|
!30298 Package the cuda operators as a dynamic link library
Merge pull request !30298 from jinjiali-kali/cuda_ops
|
4 years ago |
zong-shuai
|
891f0f87d0
|
debug
|
4 years ago |
hanhuifeng2020
|
662c51c019
|
[DynamicShape][GPU]add dynamic shape support of Concat and its backward for DCN
|
4 years ago |
i-robot
|
2f09bcab3d
|
!30358 add mul dtype register to support bool && uint_8_t
Merge pull request !30358 from zhuzhongrui/pub_master2
|
4 years ago |